Just a quick question:
Why is there a seperate static object AND a collision object? Why not make the collision object collide with rigid bodies? To me this makes no sense.
Great job by the way, please keep up the good work!
Sorry, I don’t understand your question. Why it has no sense? Have you tried build with Aligorith’s bullets?
What do you mean static? All object are rigid body’s set as active or passive . . .
@vklidu: yes the point of the script is to record bullet physics, just like you do from game engine…
download it again as I did a change this morning -wasn’t working with this recent build-
@Jonathan L if you mean integrating all simulations that would be great!
vklidu - ofcourse! I totally forgot about that “recording button” option (that creates automaticly keyframes)! Why that does the trick! Thank you!
OS X build with the new addictions and fixes: http://aligorith.blogspot.fr/2012/03/bullet-integration-small-goodies-fixes.html
Build: http://www.4shared.com/zip/fe-yAegZ/BlenderBullet_2.html
Thanks guys for such a quick updates 
you did the builds from aligorith’s tomorrow branch
@liero: thank you for update too
I played with it and it’s incredible (works like a charm with 2000 cubes at once)! However I was thinking if there’s a chance these simulations would be multithreaded? And what about creating a kind of a rope? Is it possible?
Sorry, I don’t understand your question. Why it has no sense? Have you tried build with Aligorith’s bullets?
What do you mean static? All object are rigid body’s set as active or passive . .
Yes, I was using his builds, sorry if this was confusing because I accidentally said static and not passive. I was just wondering why rigid bodies do not collide with collision objects.
This OS X build: http://www.4shared.com/zip/N86OYzaj/BlenderBullet.html comes with this recent commits:
[QUOTE=]• Thursday March 29 2012 02:42
◦ aligorith committed 4735341- Added missing notifiers (for groups) when adding/removing rigidbodies
◦ 02:22
◦ aligorith committed cfe0e09- Cosemetic code-style tweaks + rationale behind collision shape types
• Wednesday March 28 2012 16:21
◦ sergof committed a2a2705- bullet: Add null check when deleting dynamics world
◦ 16:21
◦ sergof committed 88a81cc- bullet: Use bvh-triangle mesh for passive objects
◦ 16:21
◦ sergof committed 4e29fe4- bullet: rename rbStepSimulationAdv -> rbStepSimulation, remove simple version
◦ 16:21
◦ sergof committed da9368b- bullet: Do proper cleanup when deleting rbDynamicsWorld
◦ 16:18
◦ sergof committed bc40cf1- bullet: Change the way simulation stepping works.
◦ 16:18
◦ sergof committed b7a5643- bullet: remove redundant include
◦ 16:17
◦ sergof committed 53ce987- bullet: Remove TODO comment about single vs double precision floats
◦ 16:17
◦ sergof committed b33c297- bullet: Partial fix for physics world resetting
◦ 16:17
◦ sergof committed 8cd4443- Remove axis sweep broadphase code.
◦ 16:17
◦ sergof committed ffb5ef1- bullet: Remove TODO comment about using another allocator[/QUOTE]
Check here for more details and recent updates: https://gitorious.org/~aligorith/blenderprojects/aligoriths-blender/commits/BulletDev
did someone build it for Linux?
Yeh! I just noticed we can use objects like rings. When I tried them few weeks ago they just flyied a way (in BGE too).
Speaking of magnetic, would it be possible to recreate real world magnets now that we have this not in the GE? I feel that could be handy.
Yes, geometry and/or curves as force fields. That would make possible things like this: https://vimeo.com/12545791
We do have a “Curve Guide” force field type.
It seems broken for rigid bodies though, I’ll look into fixing it when I get to do work on force fields.
It will have to wait a little though, since there are more important issues right now.